simulation of solidified microstructure and experimental...

Research ArticleSimulation of Solidified Microstructure and ExperimentalComparative Study of Twin-Roll Casting Aluminum Alloys

TengMa,1,2 Junting Zhang,3 Xiaochao Cui,1,3 and Xiaosi Sun1

1School of Materials Science and Engineering, Taiyuan University of Science and Technology, Taiyuan 030024, China2Huake Institute, Taiyuan University of Science and Technology, Taiyuan 030024, China3School of Applied Science, Taiyuan University of Science and Technology, Taiyuan 030024, China

Correspondence should be addressed to Xiaochao Cui; [email protected]

Received 12 April 2017; Accepted 18 June 2017; Published 2 October 2017

Academic Editor: Paolo Ferro

Copyright © 2017 Teng Ma et al.This is an open access article distributed under the Creative Commons Attribution License, whichpermits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

A coupledmacro-micromathematical model of twin-roll casting was established in the study.The continuous solidification processof Al-10Mg aluminum alloys in front of the nip point was numerically simulated by the cellular automaton method and thesolidification microstructure, dendritic grain radius, and secondary dendrite arm spacing were obtained through simulation topredict the mechanical property of wedge strips. In order to verify the reliability of the simulation results, the metallographicexamination and tensile tests were performed with the as-cast specimens. The results showed that the gain size, the distributioncharacteristics of various grain regions, dendrite arm spacing, and the yield strengths obtained from simulation were consistentwith experimental results.

1. Introduction

Twin-roll casting is characterized by the short process, lowenergy consumption, and lowmanufacturing cost and widelyused to produce aluminum alloy strips with goodmechanicalproperties [1–4]. The solidification structure prepared bycontinuous casting affects alloy properties as well as thequality andmechanical properties of alloy strips [5–7].There-fore, it is necessary to develop new rolling technologies anddeeply explore the relationship between the casting process,solidification structure, and metal performance.

The solidification process of strips can be simulated withthe computer simulation technology to study nucleationand grain growth and analyze solidification structure andgeometric features. The formation process of the solidifica-tion structure can be displayed directly with the simulationtechnology. Several mathematical models were developedfor the twin-roll casting process. Some micro mathematicalmodels were established to simulate the evolution process ofthe solidification microstructure in a micro region by Liu etal. [8], S. D. Chen and J. C. Chen [9], and Huang et al. [10].However, the complete solidification process of the strip or

the prediction of mechanical properties of the as-cast alloystrip was not reported. The casting-state microstructure ofaluminumalloy or other alloys had been explored in the twin-roll casting experiments [11–13]. However, the solidificationevolution was seldom explored.

According to cellular automata theory, the mathematicalmodel of the solidification structure of twin-roll continuouscasting was established in this study. With Al-10Mg alu-minum alloys as the study target, the continuous solidifica-tion process in front of the nip point was numerically simu-lated in Procast.The dendritic grain radius and the secondarydendrite arm spacing were obtained to predict the mechani-cal properties of wedge strips. Finally, the mathematic modelwas verified through metallographic examination and tensiletests.

2. Mathematic Models

2.1. Macroscopic Model. Macro temperature field is the basisfor simulatingmicrostructures. According to the heat transferprocess between the metal liquid, casting roller, and the

HindawiAdvances in Materials Science and EngineeringVolume 2017, Article ID 9304038, 7 pageshttps://doi.org/10.1155/2017/9304038

https://doi.org/10.1155/2017/9304038

2 Advances in Materials Science and Engineering

environment, Xiong et al. proposed the mathematical modelbased on unsteady heat conduction [14]:

𝜌𝑐𝑝 𝜕𝑇𝜕𝑡 = 𝜕𝜕𝑥 (𝜆𝜕𝑇𝜕𝑥) + 𝜕𝜕𝑦 (𝜆𝜕𝑇𝜕𝑦 ) + 𝜕𝜕𝑧 (𝜆𝜕𝑇𝜕𝑧 )

− 𝜌𝐿𝜕𝑓𝑠𝜕𝑡 ,(1)

where 𝜌 is alloy density; 𝑐𝑝 is a specific heat at constantpressure; 𝜆 is thermal conductivity; 𝐿 is latent heat; and 𝑓𝑠is solid fraction.

2.2. Microscopic Model

2.2.1. Nucleation Model. During the twin-roll continuouscasting, the undercooling of the liquid phase is generally farless than the undercooling required for uniform nucleationand liquid-phase surface oxidation, broken dendrites, andheterogeneous nucleation exist in the liquidmetal.Therefore,the heterogeneous nucleation model is adopted to describethe solidification process of liquid metal. At a given under-cooling Δ𝑇, the continuous nucleation model based onGaussian distribution and proposed by Gandin and Rappaz[15] is adopted:

𝑑𝑛𝑑 (Δ𝑇) = 𝑛max√2𝜋Δ𝑇0 exp[−12 (Δ𝑇 − Δ𝑇𝑁Δ𝑇0 )2] , (2)

where Δ𝑇0 is the standard deviation; 𝑛max is the maximumdensity of nuclei; andΔ𝑇𝑁 is themean nucleation undercool-ing.

2.2.2. Growth Model. Twin-roll continuous casting is a near-rapid solidification process. Therefore, the modified KGTmodel is adopted to describe the constrained dendrite tipgrowth. The dynamic effect in the dendrite growth is notsignificant [16], so compositional undercooling and cur-vature undercooling other than the kinetic degree of theundercooling of dendrite tip are considered in the model.Assuming that the equilibrium distribution coefficient ofalloy and diffusion coefficients of solute remain unchanged,the modified KGT model is given as [8]

Δ𝑇 = Δ𝑇𝑐 + Δ𝑇𝑅,Δ𝑇𝑐 = 𝑚𝑐[1 − 11 − 𝐼V (pe) (1 − 𝑘)] ,

Δ𝑇𝑅 = 2Γ𝑅 ,(3)

where 𝑐 is the solute concentration at the liquid interface; 𝑘 isthe solute partition coefficient; (pe) is the Ivantsov function;and Γ is the Gibbs-Thomson coefficient. According to thestability principle of the dendrite tip, the growth velocity of

Table 1: Chemical composition of Al-10Mg alloy.

Al Mg Si Fe Cu Zn Other components (Mn + Ni)89.70 9.90 0.11 0.18 0.01 0.03 0.01

dendrite tip V and the growing radius of dendrite tip 𝑅 can becalculated as follows:

𝑅 = 2𝜋√ Γ𝑚𝐺𝑐𝜁𝑐 − 𝐺,pe = 𝑉𝑅2𝐷𝐿 ,V2 ( 𝜋2Γ

pe2𝐷2𝐿) + V{ 𝑚𝐿𝐶0 (1 − 𝐾0)𝐷𝐿 [1 − (1 − 𝐾0) 𝐼V (pe)]} + 𝐺= 0.

(4)

The secondary dendrite spacing is an important param-eter to describe characteristics of solidification structure andcan reflect the micro segregation and other casting defects.The secondary dendrite spacing, 𝜆2, is given as [17]

𝜆2 = 5.5 (𝑀𝑡𝑓)1/3 , (5)

where

𝑡𝑓 = Δ𝑇0𝐺𝐿𝑅,

𝑀 = 𝑇𝑚Γ𝐷𝐿 ln (𝐶𝑓/𝐶0)𝑚𝐿 (𝑘 − 1) (𝐶𝑓/𝐶0) ,

(6)

where 𝐶𝑓 is the final composition of liquid metal of dendriteroot. The growth velocity of columnar crystals and dendritetip as well as the dendrite spacing in the process of twin-rollcontinuous casting can be expressed by the modified KGTmodel.

3. Experiment and Model

3.1. Experimental. The alloy material used in the study is Al-10Mg aluminum alloy (Table 1). Aluminum ingot and mag-nesium ingot were melted in an airtight furnace according tothe proportion of Al-10Mg aluminum alloy and inert gaseswere injected into the furnace to protect the melting process.The casting roller size is Ø400mm/Ø300mm; the roll gapwidth is 2–4mm; the gap between nozzle and roll is 0.10∼0.15mm. The chamber was preheated to 350∼400∘C and thetemperature was maintained. Then the twin-roll caster wasstarted and molten metal was poured into the chamber at thepouring temperature of 690∘C.The casting speed of the twin-roll caster was 1m/min.

When the aluminum alloy strip went through the rollingarea, plastic deformation occurred and primary crystal shapewas modified. In order to get initial solidification, the wedgestrip was taken out after casting and processed to the length

Advances in Materials Science and Engineering 3

Table 2: Thermal physical properties of Al-10Mg alloy.

Melting temperature (∘C) Specific heat (J⋅kgK−1) Density of liquid (kg⋅m−3) Thermal conductivity of liquid (W⋅mk−1)607 1088.6 2.55 108.9

(a) (b)

Figure 1: Twin-roll cast Al-10Mg alloy wedge-shaped plate specimen: (a) wedge-shaped plate; (b) mounting specimen.

Strip

Simulated area

Nozzle

Roll sleeve

y

x

R

Figure 2: Diagram of the twin-roll casting process.

of 15mm (Figure 1). After polishing, the strip was corrodedin 50% HF. The rolled plane and the longitudinal section ofthe sample were observed by the metallographic microscope(LeiCADM2700M).

Three groups of standard samples (10mm × 4mm ×3mm) were taken out along three directions: the horizontaldirection, rolling direction, and the angle of 45∘. The tensiletestswere carried outwith universal testingmachine to obtainmechanical properties and stress-strain curve of Al-10Mg(strain rate 𝜀 = 5 × 10−4 s−1).3.2. Geometrical Model. In the process of twin-roll continu-ous casting, the continuous nucleation of the molten metaloccurred at the meniscus of the roller when the rollerswere reversely rotated. The cast-rolled strip was thickenedconstantly in the growth of dendrite.When the stripwas closeto the nip point, the solidification process was completed andthe initial structure was formed. Therefore, the continuoussolidification process in front of the nip point is numericallysimulated in the study. The calculation area is shown inFigure 2 and the thermal physical properties of Al-10Mgalloy are shown in Table 2 [18]. The microstructure was

analyzed and the mechanical properties were predicted. Thesimulation parameters were consistent with the productionparameters.The thickness of the strip is 3mm. Figure 2 showsthe schematic diagram of the twin-roll casting process. Underthe conditions of continuous casting, the heat transfer in theroll barrel direction is far less than that in the other twodirections. Therefore, the changes in the grain morphologymainly occurred in the longitudinal section and the rolledplane of the sample, but the grain morphology was the samein the strip width direction. Therefore, in order to simplifythe model and reduce the calculation load, a certain sectionalong the roll barrel direction was selected for analysis in themodel under the assumption of adiabatic side seal. The stripwidth is 5mm in the simplified model.

4. Results and Discussion

Figures 3 and 4, respectively, show the simulated andexperimentally obtained microstructures of rolled planeand longitudinal section, which are 1mm away from theroll in the continuous casting process. Figure 5 shows theschematic diagram of sample selection for microstructure


(a) (b)

Figure 3: Solidification simulation of twin-roll cast Al-10Mg alloy wedge-shaped plate: (a) rolled plane; (b) longitudinal section.

(a) (b)

Figure 4: Solidification simulation of the cross section and lengthwise section of twin-roll cast Al-10Mg alloy wedge-shaped plate: (a) rolledplane; (b) longitudinal section.

Casting direction

Rolled plane

Longitudinal section

Figure 5: Schematic illustration of sample selection for microstruc-ture and texture analysis.

and texture analysis. On the surface of rolled plane, thesize of grains is uniform and equiaxed (Figure 3(a)). Thegrains near the casting zone are relatively larger (the leftside of Figure 3(a)), whereas the grains near the rollingzone are relatively smaller (the right side of Figure 3(a)). Inthe simulation results, the grain size is 3.2 × 10−8m2–4.9× 10−7m2, whereas metallographic test results indicate thatthe grain size is 8.0 × 10−8m2–8.1 × 10−7m2. The aboveresults prove that the calculated results are well consistentwith experimental results. The simulated grain size is a littlesmaller than the result by metallographic test because of theassumption made in the theoretical models. In addition, thesolidification process is a complicated physical process and itis difficult to exactly determine the changing rule of thermalproperties with temperature. In Figure 3(b), the solidification

structure of longitudinal section from the roll surface intothe strip center is, respectively, composed of the fine grainzone, columnar grain zone, and equiaxed crystal zone. Thepredicted results including the grain shape and grain size areconsistent with the results of metallographic tests.

Figure 6(a) shows the fraction solid distribution ofmolten metal at the steady state. Figures 6(b) and 6(c),respectively, show primary dendritic radius distribution andsecondary dendrite arm spacing distribution. The primarydendritic radius and the secondary dendrite arm spacingare important parameters for estimating the solidificationstructure and can reflect various roll casting defects. Thesolidification process of the solution reaches a stable stateafter the casting process becomes stable. The primary den-dritic radius in the front of columnar grain growth zonereflects the growth condition of columnar grain growth. Thesmaller radius indicates the more obvious growth tendency.Figure 6(b) shows that the primary dendritic radius of stripsurface is smaller than that of strip canter. As shown inFigure 6(c), the secondary dendrite arm spacing is graduallydecreased from the outer zone to the inner zone alongthe thickness direction and is gradually increased from theouter zone to the inner zone along the rolling directionwith the solidification. The largest secondary dendrite armspacing occurs near the nip point and is about 19.55 𝜇m.Themeasured result (23𝜇m) in Figure 4(a) is basically consistent


1.0000.9330.8670.8000.7330.6670.6000.5330.4670.4000.3330.2670.2000.1330.0670.000

Fraction solid

(a)

1133.51057.9982.4906.8831.2755.7680.1604.5529.0453.4377.8302.3226.7151.175.60.0

Dendritic grain radius (micron)

(b)

19.55

18.25

16.95

15.64

14.34

13.04

11.73

10.43

9.13

7.82

6.52

5.21

3.91

2.61

1.30

0.00

Dendritic secondary arm spacing (micron)

(c)

Figure 6: Solid fraction graph, dendritic grain radius graph, and secondary dendrite arm spacing graph of twin-roll cast Al-10Mg alloywedge-shaped plate: (a) solid fraction graph; (b) dendritic grain radius graph; (c) secondary dendrite arm spacing graph.

with the simulation result, indicating that the model isapplicable to predict the dendrite arm spacing.

Roll casting solidification is a sub-rapid solidificationprocess. Compared with the traditional casting structures,the strip structure possesses the smaller grain size, lowersegregation degree, and excellent mechanical properties.In this paper, the mechanical properties of the strip werepredicted based on the solidification structure obtained bythe mathematic model of the casting process. The predictedyield strength of twin-roll cast Al-10Mg strip was 158Mpa,which was a little higher than the experimental result,142Mpa (Figure 7), due to the defects such as segregation,gas cavity, and the heterogeneity of grain size. The experi-mental results of uniaxial tensile mechanical properties ofAl-10Mg aluminum alloy casting strip at room temperatureare shown in Table 3. The corresponding engineering stress-strain curve is shown in Figure 8. According to the aboveresults, the Al-10Mg casting strip shows the goodmechanical

Table 3: Tensile properties of twin-roll cast Al-10Mg plate.

Orientation 𝜎0.2/MPa 𝜎𝑏/MPa 𝛿0∘ 142.9 207.2 8.690∘ 121.1 183.3 7.445∘ 108.6 159.4 5.3

properties, obvious yield phenomenon, and higher elonga-tion.

5. Conclusions

(1) A mathematical model of the solidification process ontwin-roll continuous casting thin strip was developed basedon the cellular automata (CA) theory. The continuous solid-ification process in front of the nip point was numericallysimulated by Procast. The solidification microstructure, the


Yield strength (MPa)

143.4

154.4

165.4

132.3

121.3

110.3

77.2

55.1

99.3

66.2

44.1

33.1

22.1

11.0

88.2

0.0

Figure 7: Simulation results of the yield strength of Al-10Mg alloyplate.

0∘

45∘

90∘

0

20

40

60

80

100

120

140

160

180

200

220

Stre

ss (M

pa)

1 2 3 4 5 6 7 8 9 100

Strain (%)

Figure 8: Engineering stress-strain curves of twin-roll cast Al-10Mgalloy plate at room temperature.

grains size, the dendritic grain radius, and the secondary den-drite arm spacing were obtained to predict the mechanicalproperties of the wedge strip.(2)The simulation results were verified bymetallographicexamination and tensile tests. The simulation results of themicrostructure on the wedge strip rolled plane and cross sec-tion, whichwas 1mmaway from the roll, were consistent withthe experimental results. The simulation results of dendriticgrain radius and the secondary dendrite arm spacing were,respectively, 1133 𝜇m and 16 𝜇m, which were consistent withthe experimental results. The predicted yield strength of thealuminum alloy strip was 158MPa, which was consistent withthe experimental results (142.9MPa).(3) The solidification process of twin-roll cast alloywas described accurately with the established mathematical

model. The model can be used to predict the solidificationstructure and mechanical properties of the strip underdifferent process parameters and solidification parameters.

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper.

Acknowledgments

This work was supported by the National Natural Sci-ence Foundation of China (51574171 and 51641406) and theNatural Science Foundation of Shanxi Province of China(2015011002).

References

[1] P. D. Ding, B. Jing, C. M. Yang, and L. Fang, “Development andthought of thin-strip continuous casting,” The Chinese Journalof Nonferrous Metals, vol. 14, no. S1, pp. s192–s196, 2004.

[2] H. S. Di, P. W. Bao, Y. C. Miao, G. D. Wang, and X. H. Liu,“Experimental study on the twin roll strip casting and analysisof processing stability,” Journal of Northeastern University (Nat-ural Science), vol. 21, no. 3, pp. 274–277, 2000.

[3] Y. Birol, “Analysis of macro segregation in twin-roll cast alu-minium strips via solidification curves,” Journal of Alloys andCompounds, vol. 486, no. 1-2, pp. 168–172, 2009.

[4] C. Gras, M. Meredith, and J. D. Hunt, “Microstructure andtexture evolution after twin roll casting and subsequent coldrolling of Al-Mg-Mn aluminium alloys,” Journal of MaterialsProcessing Technology, vol. 169, no. 2, pp. 156–163, 2005.

[5] H.-L. Mao, J.-H. Wang, G. Yang, Q.-N. Shi, J.-H. Yi, and L.-W.Chen, “Characterization of microstructure of 1235 aluminumalloy sheet produced by twin roll casting,” The Chinese Journalof Nonferrous Metals, vol. 24, no. 4, pp. 863–869, 2014.

[6] M. C. Flemings and R. W. Cahn, “Organization and trendsin materials science and engineering education in the US andEurope,” Acta Materialia, vol. 48, no. 1, pp. 371–383, 2000.

[7] K. Shibuya, F. Kogiku, M. Yukumoto, S. Miyake, M. Ozawa, andT. Kan, “Development of a rapid solidification process with adouble-roller method,” Materials Science and Engineering, vol.98, pp. 25–28, 1988.

[8] X.-B. Liu, Q.-Y. Xu, T. Jing, and B.-C. Liu, “Microstructureof aluminum twin-roll casting based on Cellular Automation,”Transactions of Nonferrous Metals Society of China (EnglishEdition), vol. 18, no. 4, pp. 944–948, 2008.

[9] S. D. Chen and J. C. Chen, “Micro-model of simulation on twin-roll continuous casting thin strip solidification structure,” RareMetal Materials and Engineering, vol. 42, no. 1, pp. 14–18, 2013.

[10] F. Huang, H. S. Di, and G. S. Wang, “Numerical simulationon solidification microstructure evolution in twin-roll castingprocess,” Casting. Forging. Welding, vol. 40, no. 5, pp. 57–60,2011.

[11] J.-H. Cho, H.-W. Kim, C.-Y. Lim, and S.-B. Kang, “Microstruc-ture and mechanical properties of Al-Si-Mg alloys fabricatedby twin roll casting and subsequent symmetric and asymmetricrolling,” Metals and Materials International, vol. 20, no. 4, pp.647–652, 2014.

[12] H. Harada, S. Nishida, E. Masaki, and H. Watari, “Casting ofhigh-aluminum-content Mg alloys strip by a horizontal twin-roll caster,” Metallurgical and Materials Transactions B: Process


Metallurgy and Materials Processing Science, vol. 45, no. 2, pp.427–437, 2014.

[13] S. J. Yao, X. K. Wang, J. Xu, N. M. Han, and S. K Guan, “Micro-structure analysis of cast rolling 1235 aluminum alloy plates,”Light Alloy Fabrication Technology, vol. 42, no. 4, pp. 30–34,2014.

[14] S. M. Xiong, Q. Y. Xu, and J. W. Kang,Modeling and SimulationTechnology in Casting Process, China Machine Press, Beijing,China, 2004.

[15] Ch.-A. Gandin and M. Rappaz, “A coupled finite element-cellular automaton model for the prediction of dendritic grainstructures in solidification processes,” Acta Metallurgica etMaterialia, vol. 42, no. 7, pp. 2233–2246, 1994.

[16] M. B. Yang, F. S. Pan, X. D. Peng, and P. D. Ding, “Macroscopicand microscopic coupling mathematical model,” Journal ofChongqing University (Natural Science Edition), vol. 28, no. 5,pp. 39–44, 2005.

[17] B. Y. Sun, Direct Strip Casting of Metals and Alloys Processing,Microstructure and Properties, National Defense Industry Press,Beijing, China, 2014.

[18] B. Y. Hang, C. G. Li, L. K. Shi, G. Z. Qiu, and T. Y. Zuo, Chi-nese Materials Dictionary Book 4, China Material EngineeringCanon, 2008.

Submit your manuscripts athttps://www.hindawi.com

ScientificaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

CorrosionInternational Journal of

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Polymer ScienceInternational Journal of



CeramicsJournal of


CompositesJournal of

NanoparticlesJournal of



International Journal of

Biomaterials


NanoscienceJournal of

TextilesHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Journal of

NanotechnologyHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

CrystallographyJournal of


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


CoatingsJournal of

Advances in

Materials Science and EngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Smart Materials Research



MetallurgyJournal of


BioMed Research International

MaterialsJournal of


simulation of solidified microstructure and experimental...

Documents